Mainz 2022 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

P: Fachverband Plasmaphysik

P 9: Poster I

P 9.12: Poster

Tuesday, March 29, 2022, 16:00–17:30, P

Physics-informed neural network of the ideal-MHD model in Wendelstein 7-X configurations — •Andrea Merlo, Daniel Böckenhoff, Jonathan Schilling, Samuel Aaron Lazerson, Thomas Sunn Pedersen, and the W7-X Team — Max-Planck-Institute for Plasma Physics, 17491 Greifswald, Germany

In magnetic confinement fusion research, the achievement of high plasma pressure is key to reaching the goal of net energy production. The magnetohydrodynamic (MHD) framework is used to self-consistently calculate the effects the plasma pressure induces on the magnetic field used to confine the plasma. In stellarators (e.g., Wendelstein 7-X), the confining field is inherently 3D, making MHD calculations costly to compute (O( 1 ) CPUh). In this work, we describe a Physics-Informed Neural Network which has been trained not only to reproduce ground-truth magnetic equilibria computed with a traditional solver (e.g., VMEC), but also to satisfy the flux surface averaged pressure balance equation characterizing ideal-MHD. The NN model is benchmarked against VMEC on a set of W7-X magnetic configurations at finite volume averaged beta, and the computation of a set of representative physics quantities of interests (e.g., magnetic well) is used to validate the model use in addressing magnetic equilibrium dependent physics questions.